Impact of Drivetrain on Wind Farm VAR Control - Upwind

More documents

Recommendations

Info

UPWIND PI control settings tuned based on having all WTGs in service PI control settings tuned based on having half <strong>of</strong> the WTGs in service Figure 4-18. System response with VAR control for different PI controller settings with half <strong>of</strong> the WTGs in service (brushless excitation, SCR=5). 4.2.3 Test Cases and Conclusions The same test cases conducted with the static excitation system were investigated with the brushless-type. The results are summarized in Table 4-10. Generally, when the control parameters are within the excitation system field voltage capability and the WTG voltage limits, the desired response can be obtained. But with higher system SCR, the minimum limit <strong>of</strong> the WTG voltage or the limit <strong>of</strong> the field voltage can be reached. This can happen with lower WTG initial voltage which implies lower margin to the minimum field voltage limit. The following conclusions can be drawn: • Scenario 1: With full power wind farm and all WTGs in service, a reasonable response time (a few seconds) for the POI voltage was obtained for system SCRs 5 and 3. The settling time was long for SCR 20. This was mainly due to the intentionally larger time response <strong>of</strong> the VAR control to prevent voltage collapse for scenario 3. • Scenario 2: With half power wind farm and all WTGs in service, a reasonable response was also obtained for system SCRs 5 and 3. The settling time was longer for SCR 20. Deliverable D5.9.1 56
UPWIND • Scenario 3: With half power wind farm and half <strong>of</strong> WTGs in service, a reasonable response was obtained with system SCRs 5 and 3. With system SCR 20, a voltage collapse occurred when Ki was determined using a 3 seconds VAR control time constant. This is shown in Figure 4-19. The WTG terminal voltage reacts on a fast time scale the excitation system reacts relatively slowly, and so could not regulate the WTG voltage to a stable condition. In this case, the VAR control time response was slowed down where the VAR control time constant was increased and set to 5 seconds. This is achieved by having a lower value <strong>of</strong> the integral gain Ki <strong>of</strong> the PI controller while maintaining the value <strong>of</strong> Kp/Ki to cancel out the excitation system time constant. However in steady state, the POI voltage could not settle to the desired value because the WTG terminal voltage limit was reached. In order to obtain the required performance with the wind VAR control, within the excitation system capability with high SCR, coordination could be made with other passive elements. • Scenario 4: With VAR control settings for system SCR 20 applied for system SCRs <strong>of</strong> 5 and 3, the VAR control response is faster and has a slight overshoot because the control settings were based on higher Ki making the POI voltage settle more quickly. When the control settings for system SCR 3 were applied for system SCR 20, the VAR control was slower due to the lower Ki value determined with SCR 3. The POI voltage does not settle within the 20-second window. • Scenario 5: With higher WTG initial voltage with system SCR 20 in scenario 2, the WTG voltage has more margin before reaching voltage limits. The WTG voltage minimum limit was not reached and the settling time was slightly less than that in case 4 with lower overshoot. The implementation <strong>of</strong> only WTG voltage control did not consider the WTG reactive power capability. This limitation is relevant in cases 4 and 7 where the final WTG reactive power exceeded the unit minimum limit. The test cases confirmed the sensitivity <strong>of</strong> the system response with higher system SCR. The brushless excitation system response might impose constraints on the VAR control. Coordination with passive elements can provide for more margin to the WTG voltage limits. Figure 4-19. A voltage collapse case with half <strong>of</strong> WTGs for system SCR 20. Deliverable D5.9.1 57
Page 1 and 2:
Project UpWind Contract No.: 019945
Page 3 and 4:
UPWIND Contents Executive Summary .
Page 5 and 6: UPWIND Figure 4-21. System response
Page 7 and 8: UPWIND Glossary DFIG Doubly Fed Ind
Page 9 and 10: UPWIND conventional synchronous mac
Page 11 and 12: UPWIND dynamic model basis as in th
Page 13 and 14: UPWIND available wind power is less
Page 15 and 16: UPWIND Table 2-2. Wind farm system
Page 17 and 18: UPWIND 3. Wind Farm VAR Control for
Page 19 and 20: UPWIND 3.1.3 Wind Farm VAR Control
Page 21 and 22: UPWIND SCR:20 SCR:5 y-axis (min) y-
Page 23 and 24: UPWIND SCR:20 SCR:5 SCR:3 Figure 3-
Page 27 and 28: UPWIND +/- 10% of
Page 29 and 30: UPWIND Scenario Case SCR MVA base W
Page 31 and 32: UPWIND Table 4-1. Parameters <stron
Page 33 and 34: UPWIND Magnitude (dB) Phase (deg) 2
Page 37 and 38: UPWIND 20 3 0.2 1.65 3.3 5 3 0.49 0
Page 39 and 40: UPWIND 4.1.5 Results Summary and Co
Page 41 and 42: UPWIND Table 4-4. Summary results <
Page 43 and 44: UPWIND Given knowledge of</
Page 45 and 46: UPWIND reactance (Xgrid). The ratio
Page 49 and 50: UPWIND Table 4-7. Summary results <
Page 51 and 52: UPWIND Magnitude (dB) Phase (deg) 6
Page 55: UPWIND Vreg_ref + - K K i p ( 1+ s)
Page 59 and 60: UPWIND 4.2.4 Wind Farm VAR Control
Page 63 and 64: UPWIND 4.2.5 Test Cases and Results
Page 65 and 66: UPWIND 5. Summary Wind farms can be
show all

Impact of Drivetrain on Wind Farm VAR Control - Upwind

Create successful ePaper yourself

Delete template?

Save as template?